Generative AI has quickly become a standard productivity tool, but insurers may need to move beyond general-purpose applications if they want to gain a lasting competitive advantage.

Research from Stanford found that generative AI can significantly improve productivity on digital tasks, including drafting emails, summarising reports and developing internal tools. Earnix argues that as access to the same general-purpose AI models becomes widespread, productivity gains alone are unlikely to differentiate insurers.

The next opportunity lies in vertical AI designed around specific industries and business processes. Rather than simply helping individual employees work faster, these systems can connect decision-making across an organisation and incorporate industry-specific requirements.

AI is already used across insurance, from underwriting and pricing to claims and customer engagement. Underwriters can use models to assess assets, geographic and regulatory factors and climate data, while claims teams use automated anomaly detection to identify potentially fraudulent activity. Customer-facing teams can also use generative AI to understand policyholders’ circumstances and generate tailored responses.

As these applications become more common, their availability is becoming less of a differentiator. The bigger challenge is fragmentation between the different AI systems insurers use across their businesses.

Many insurers rely on a combination of internally developed and third-party AI tools that operate independently. This can leave data and insights trapped within individual departments, potentially creating inconsistencies between functions such as claims, customer service, underwriting and pricing.

The answer, according to Earnix, is what it describes as insurance-native AI, with platforms designed around the industry’s regulatory, operational and data requirements.

Source link